The present invention relates to an apparatus that has a plurality of slots and makes channel estimation (propagation path estimation) of data symbols from pilot symbols in a combined symbol sequence including the data symbols and pilot symbols, and a CDMA (Code Division Multiple Access) receiver and CDMA transmitter with the apparatus.
In a mobile communications environment, amplitude and phase fluctuations in a traffic channel can occur because of Raleigh fading due to changes in the relative location between a mobile station and a base station. Thus, in a conventional phase modulation scheme that transmits data (information) by the phase of a carrier, it is common for a transmitting side to carry out differential encoding of transmitted data for impressing the data on relative phases of neighboring symbols, and for a receiving side to discriminate and decide the data by differential detection.
However, since the transmitted data is subjected to the differential encoding as mentioned above, a one-bit error in a radio section appears as a two-bit error in the differential detection, thereby increasing the receiving error rate by 3 dB in terms of the SNIR (Signal-to-Noise Interference power Ratio) as compared with coherent detection like binary phase-shift keyed modulation (BPSK modulation).
On the other hand, although absolute coherent detection, which discriminates and decides the phase of a received signal using the absolute phase of each data symbol, has a highly efficient receiving characteristic, it is difficult under the Raleigh fading environment to decide the absolute phase of the reception.
In regard to this matter, reference 1, Seiichi Sampei and Terumi Sunaga, xe2x80x9cRaleigh Fading Compensation for QAM in Land Mobile Radio Communicationxe2x80x9d, IEEE Trans. Vehicular Technol., VT-42, No. 2, May 1993 proposes a method of estimating and compensating for fading distortion using pilot symbols that are inserted in data symbols at fixed intervals, and have known phases. In the reference 1, a pilot symbol is inserted at every several data symbols so as to carry out the channel estimation based on the received phase of the pilot symbol. More specifically, using the pilot symbols before and after the data symbol section, the method measures the amplitude and phase of a received signal of each path of each user, and estimates and compensates for channel fluctuations in the data symbol section by interpolating the values measured.
On the other hand, reference 2, Hidehiro Ando et al., xe2x80x9cChannel Estimation Filter Using Time-Multiplexed Pilot Channel for Coherent RAKE Combining in DS-CDMA Mobile Radioxe2x80x9d, IEICE Trans. Commun. Vol. 81-B, No. 7 July 1998 proposes a method of carrying out channel estimation with higher accuracy by performing the channel estimation using more pilot symbols.
FIG. 11 illustrates a channel estimation method disclosed in the reference 2. This method carries out transmission power control on a slot by slot basis to follow instantaneous Raleigh fluctuations. Accordingly, as shown in FIG. 11, the amplitude (power) of a combined symbol sequence consisting of the data symbols and pilot symbols varies slot by slot, and its phase also varies slightly due to the operation of an amplifier in transmission. Such transmission power control enables a reverse channel of the DS-CDMA (Direct Sequence CDMA) to maintain the SNIR against interference signals due to cross-correlation from other users.
The channel estimation of data symbols is performed using the pilot symbols inserted into the data symbols at fixed intervals. More specifically, it obtains its channel estimates {tilde over ("xgr")} by averaging (coherently adding) pilot symbols {circumflex over ("xgr")} (estimated complex fading envelope) in multiple slots before and after the slot to which the data symbols to be estimated belong, and then by summing the averages {circumflex over ({overscore (86 )})} weighted by weighting factors a. Highly accurate channel estimation is carried out in this manner.
With such channel estimation using many pilot symbols belonging to different slots, this method can achieve the channel estimation at higher accuracy. This is because although the power of the pilot symbols fluctuates in the multiple slots, and channel estimation error takes place due to the power fluctuations, an effect of reduction in thermal noise and interference signals obtained by using pilot symbols in many slots is greater than the channel estimation error.
However, it is difficult for the method of the reference 2 to achieve the channel estimation with further accuracy because it considers the channel fluctuations in the individual slots are small, and obtains the channel estimates {circumflex over ("xgr")} using the same weighting factor a for all the data symbols in each slot.
For example, as shown in FIG. 11, this method uses, even for the (mxe2x88x92A)th data symbol or the (m+B)th data symbol in the nth slot, where A and B are natural numbers, the same weighting factors a(0), a(1) and the like to obtain their channel estimates {tilde over ("xgr")} (n).
However, with regard to the (mxe2x88x92A)th data symbol, it will be reasonable to assign a greatest weight to the pilot symbols in the nth slot because they are closest (in time) to the (mxe2x88x92A)th data symbol, and. Hence best reflect the channel state at the time the data symbol is transmitted.
In contrast with this, with regard to the (m+B)th data symbol, it will be reasonable to assign a greatest weight to the pilot symbols in the (n+1)th slot because they are closest (in time) to the (m+B)th data symbol, and hence best reflect the channel state at the time the data symbol is transmitted.
Thus, the channel estimates should be obtained by assigning proper weighting factors to individual data symbols even though they belong to the same slot.
FIG. 12 illustrates an example of received envelope fluctuations due to fading. Points 1205, 1210, 1215, 1220 and 1225 indicate in fast fading the values of a received envelope at fixed time intervals. Points 1255, 1260, 1265 and 1270 indicate in slow fading the values of a received envelope at the same fixed time intervals.
The received envelope fluctuations are greater in the fast fading than in the slow fading. Accordingly, it is important especially in the fast fading to carry out the highly accurate channel estimation by assigning proper weighting factors to individual data symbols even though they belong to the same slot.
The present invention is implemented to solve the foregoing problems. It is therefore an object of the present invention to achieve highly accurate channel estimation by obtaining highly accurate channel estimates by assigning appropriate weighting factors to individual data symbols in the same slot, and by calculating a sum of appropriately weighted pilot symbols in respective slots before and after the slot the data symbols belong to, when carrying out the channel estimation of the data symbols.
The highly accurate channel estimation and compensation for channel fluctuations in the data symbols based on the channel estimation make it possible for the absolute coherent detection to decide the absolute phase of each data symbol even in the Raleigh fading environment, which can reduce the SNIR for achieving desired receiving quality (receiving error rate). This can reduce the transmission power, and increase the capacity of a system in terms of the number of simultaneous subscribers.
In order to accomplish the object aforementioned, according to the invention as claimed in claim 1, a channel estimation unit for obtaining channel estimates of data symbols from pilot symbols in a combined symbol sequence which has a plurality of slots and includes the data symbols and the pilot symbols, comprises:
means for locating the pilot symbols in the combined symbol sequence;
means for generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result; and
means for obtaining the channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks,
wherein a magnitude of weighting differs between at least two data symbols in each slot.
According to the invention as claimed in claim 2, a CDMA receiver which receives a combined symbol sequence that is spread, has a plurality of slots, and includes data symbols and pilot symbols, and which generates a data sequence, comprises:
means for receiving the spread combined symbol sequence;
means for generating a combined symbol sequence by despreading the spread combined symbol sequence;
means for locating the pilot symbols in the combined symbol sequence;
means for generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result;
means for obtaining channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks;
means for obtaining a data symbol sequence by eliminating the pilot symbols from the combined symbol sequence in accordance with the located result;
means for compensating for channel fluctuations in the data symbol sequence by using the channel estimates of the data symbols; and
means for generating the data sequence by demodulating the data symbol sequence compensated for,
wherein a magnitude of weighting differs between at least two data symbols in each slot.
According to the invention as claimed in claim 3, a CDMA transceiver including a transmitting processor and a receiving processor, comprises:
means for generating a data symbol sequence by modulating a data sequence;
means for generating a combined symbol sequence by inserting pilot symbols into the data symbol sequence;
means for generating a spread combined symbol sequence by spreading the combined symbol sequence; and
means for transmitting the spread combined symbol sequence,
wherein the spread combined symbol sequence to be transmitted has a plurality of slots, and the receiving processor comprises:
means for receiving the spread combined symbol sequence;
means for generating the combined symbol sequence by despreading the spread combined symbol sequence;
means for locating the pilot symbols in the combined symbol sequence;
means for generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result;
means for obtaining channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks;
means for obtaining a data symbol sequence by eliminating the pilot symbols from the combined symbol sequence in accordance with the located result;
means for compensating for channel fluctuations in the data symbol sequence by using the channel estimates of the data symbols; and
means for generating the data sequence by modulating the data symbol sequence compensated for,
wherein a magnitude of weighting differs between at least two data symbols in each slot.
According to the invention as claimed in claim 4, in the CDMA transceiver as claimed in claim 3, he transmitting processor further comprises means or inserting into the data symbol sequence a power control symbol sequence for controlling power of the data symbols and pilot symbols.
According to the invention as claimed in claim 5, in the CDMA transceiver as claimed in claim 4, the receiving processor further comprises means for measuring from the pilot symbols a signal-to-noise and interference power ratio, and for generating the power control symbol sequence from the signal-to-noise and interference power ratio.
According to the invention as claimed in claim 6, in the CDMA transceiver as claimed in any one of claims 3-5, the receiving processor further comprises means for extracting, from the data symbol sequence compensated for, the power control symbol sequence for controlling power of the data symbols and pilot symbols, and the means for transmitting the spread combined symbol sequence transmits the spread combined symbol sequence in accordance with he power control symbol sequence.
According to the invention as claimed in claim 7, in the equipment as claimed in any one of claims 1-6, the power of the data symbols and pilot symbols is controlled on a slot by slot basis.
According to the invention as claimed in claim 8, in the equipment as claimed in any one of claims 1-7, the number of data symbols included in each slot of the combined symbol sequence is the same, and the number of pilot symbols included in each slot of the combined symbol sequence is the same.
According to the invention as claimed in claim 9, in the equipment as claimed in any one of claims 1-8, the pilot blocks each consist of all the pilot symbols in each slot.
According to the invention as claimed in claim 10, in the equipment as claimed in any one of claims 1-9, when obtaining the channel estimates of the data symbols in an nth slot in the combined symbol sequence, where n is an integer, the pilot blocks are generated from (nxe2x88x92K+1) th slot to (n+K) th slot in the combined symbol sequence, where K is a natural number.
According to the invention as claimed in claim 11, in the equipment as claimed in any one of claims 1-10, the pilot blocks closer to the data symbol with which the channel estimate is to be obtained have a greater weight.
According to the invention as claimed in claim 12, a channel estimation method of obtaining channel estimates of data symbols from pilot symbols in a combined symbol sequence which has a plurality of slots and includes the data symbols and the pilot symbols, comprises the steps of:
locating the pilot symbols in the combined symbol sequence;
generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result; and
obtaining the channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks, wherein a magnitude of weighting differs between at least two data symbols in each slot.
According to the invention as claimed in claim 13, a CDMA receiving method of generating a data sequence by receiving a combined symbol sequence that has a plurality of slots, includes data symbols and pilot symbols, and is spread, comprises the steps of:
receiving the spread combined symbol sequence;
generating the combined symbol sequence by despreading the spread combined symbol sequence;
locating the pilot symbols in the combined symbol sequence;
generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result;
obtaining channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks;
obtaining a data symbol sequence by eliminating the pilot symbols from the combined symbol sequence in accordance with the located result;
compensating for channel fluctuations in the data symbol sequence by using the channel estimates of the data symbols; and
generating the data sequence by demodulating the data symbol sequence compensated for,
wherein a magnitude of weighting differs between at least two data symbols in each slot.
According to the invention as claimed in claim 14, a CDMA transmitting and receiving method comprises the steps of:
on a transmitting side,
generating a data symbol sequence by modulating a data sequence;
generating a combined symbol sequence by inserting pilot symbols into the data symbol sequence;
generating a spread combined symbol sequence by spreading the combined symbol sequence; and
transmitting the spread combined symbol sequence,
wherein the spread combined symbol sequence to be transmitted has a plurality of slots, and
on a receiving side,
receiving the spread combined symbol sequence;
generating the combined symbol sequence by despreading the spread combined symbol sequence;
locating the pilot symbols in the combined symbol sequence;
generating pilot blocks by extracting the pilot symbols from two or more slots in the combined symbol sequence in accordance with a located result;
obtaining channel estimates of the data symbols by calculating a weighted sum of averages of the pilot symbols in the individual pilot blocks;
obtaining a data symbol sequence by eliminating the pilot symbols from the combined symbol sequence in accordance with the located result; compensating for channel fluctuations in the data symbol sequence by using the channel estimates of the data symbols; and
generating the data sequence by demodulating the data symbol sequence compensated for,
wherein a magnitude of weighting differs between at least two data symbols in each slot.